AI for Economic Opportunity Demo Day with OpenAI Academy and GitLab Foundation

Hi everyone, it is such a pleasure to be here with you. I'm Ellie Bertani, the President and CEO of GitLab Foundation.

I am absolutely thrilled to welcome all of you to our AI Demo Day, whether you're joining us here in person or remotely across the globe live-streaming.

Today is a celebration of bold ideas in service of a shared mission, showcasing how AI, this incredibly powerful tool, can be used to unlock economic opportunity and boost incomes for people struggling to improve their prospects for themselves, their families, and their communities.

At the GitLab Foundation, we believe philanthropy and the world at large is at a crossroads. In a world where technology, and in particular artificial intelligence, is driving incredibly exciting opportunities to drive social impact, and yet millions face persistent barriers to prosperity, we believe that philanthropy must embrace transparency, agility, collaborations, and an outcomes orientation if it aims to create meaningful change. And we must learn to do more with less.

Our approach begins with a single powerful question, how can we maximize the impact of every dollar that we give?

remarkable teams you'll hear from today, all members of our second AI for Economic Mobility cohort will inspire you by showcasing exactly how technology can drive this kind of impact through cost-effective, disruptive, creative innovation. We're proud to support these organizations and their work to leverage AI to make paths to upward mobility, including training, education, employment, and more, more accessible, effective, and efficient.

They're not just envisioning a brighter future, they're creating it.

Today's event is made possible by our incredible partners. A heartfelt thank you to OpenAI and the Ballmer Group who partnered with us to make this cohort happen, but also and also to every, mm, please. But in addition and also to every funder joining us today, investing in these teams and other similar solutions, your willingness to take risks on these transformative, early stage ideas is more important than ever.

And to help us get started, please welcome Sid Sijbrandij, founder of GitLab and board chair of the GitLab Foundation.

Thank you.

Thank you, Ellie, and thanks. I'm really honored to be here today. I still remember vividly three years ago when I asked Ellie to,

lead the GitLab Foundation, and we thought about how we could do the most good for the most people. And we said, millions of people need the economic mobility to start earning more. And this is Silicon Valley. We wanted to be ambitious, and we said, we want to have the biggest impact for our buck. We want at least 100x return on every dollar we invest for other people. And it felt impossible. I remember talking, maybe 50, 50 would already be great, but we aim for 100. And I couldn't be more proud of how they're doing that, what they're doing, but also how they're doing that. They are achieving.

that with radical transparency and agility. If you want an example of the radical transparency, go to their website. Every project has its returns, and they're being really rigorous and honest about what those returns are.

You want an example of the agility, look in the room today. This program didn't exist. I didn't think of it. The GitLab Foundation with Ellie thought of starting this. They were one of the first movements in the era of AI. And I'm really thankful for everyone who partnered with us, especially OpenAI and the Ballmer Foundation, but also so many of you here in the room. Thanks for making this happen.

We couldn't be more excited about what's ahead. The best thing is the results we're going to be.

be able to achieve for people. The first two years of the GitLab Foundation, they were able to do 164x. They greatly exceeded what we hoped for. And that means we're increasing the lives of people with $1.3 billion in extra income, and that's touching 220,000 people. And I... I'm so excited to see all the initiatives today, I hope you'll see that we can have great impact in the era of AI, and I'm really grateful for the GitLab Foundation organizing that.

And thank you all for being here today.

Thank you very much.

Thank you.

Thank you.

Thank you, Sid, for those overwhelmingly positive comments. Now to dive deeper into today's theme of AI and economic opportunity, I'm very excited to welcome to the stage someone who sits at the intersection of innovation, policy, and economic transformation, Dr. Ronnie Chatterji, Chief Economist of OpenAI, where he's led the efforts to understand and shape how AI impacts jobs, productivity, and inclusive economic growth. Please give him a warm welcome.

How's everyone doing today?

A little more enthusiasm, come on, good guy. This has been an amazing experience. I want to welcome our panelists here in a second and do a round table discussion with them because I think they're really the reason that you're here. But I just want to thank everyone for being here today and for all the energy you're putting into using AI and other technologies to promote social good and make an impact on people's lives.

The reason I joined OpenAI was because our mission is about how to use technology to benefit all of humanity. I studied economics for most of my life and did my PhD because I thought that I could use economics to make a difference. But so many times we have new technologies, we've fallen short, honestly, of

bringing everyone along. We've created digital divides, we've created boundaries, and as much as we've talked about it, whether it's in the ivory tower, or among philanthropies, or businesses, it's really hard to close those gaps and those disparities. And when you think about AI, we're facing a lot of the same issues that I came up with when we were thinking about the internet and access to broadband, which is that if we get more people access, we'd have more economic opportunity to go around. But if we didn't do that, we'd reinforce existing inequality, existing structures that hold people back. And it was really up to all of us to work together to try to figure out if we could solve these issues as the internet, social media, smartphones, all those technologies rolled out.

We're kind of in one of those moments in AI right now.

And I feel like the reason I joined OpenAI is there's tremendous energy here to try to address those issues. But I wanna say that we need your help.

You know, it's interesting, a lot of people come here to this room and they think that we're gonna have the answers on all these issues. And I'm very much here to tell you that we have a lot of questions for you.

When I look at the organizations in the room, and I look at the cohort last year, and I'm gonna welcome Rose and Jared up here in a second to talk about their work, you guys are providing the solutions, the answers to the questions that we're looking for.

And I think that we have to find a way to work together in partnership with the technological capabilities, the organizational capacity of OpenAI, and all the work that the foundations and the organizations that they're funding.

are doing in the audience. There's two areas I just wanna highlight that I'm thinking a lot about that I hope you'll join me in problem solving around. A lot of you are already working on these issues. Rose and Jared represent two people who are working on this as well.

First is, you know, I took this job because everyone was warning me that AI was gonna have a really negative impact on the job market. I reflected on my life studying economics and I observed that a lot of times new technologies actually end up complimenting workers, not substituting for them. I looked at the electricity revolution, the steam engine, the internet, and the story of substitution wasn't necessarily the one that was coming up in all those cases. If you look at the early studies of AI, like rigorous studies.

randomized controlled trials, the types of things you guys fund, we are now seeing much more complement than substitute. The idea is that if you can use AI to make yourself more productive, it frees yourself up to do other tasks.

For a lot of us, it means that the tasks we do at our job that we don't actually like to do that much, we can offset to the machine, and we can spend more time on the higher value and sometimes higher margin tasks that we wish to do. For a lot of people, that's working right now. But it is an open question whether that will continue to be the case.

We have to make sure that we design training programs, skilling programs, and we keep humans in the loop in terms of autonomy over their own jobs to make sure that we're moving together into the future, not just with machines ahead of humans. When you think about this-

A big part of my job is trying to figure out what indicators and forecasts to put out there. I can't solve all the problems in this role. What I can do is try to put data and evidence out there to help you make better decisions. And I hope that one of the things we can do coming out of this event is work together to try to develop what those indicators are, what those metrics are, what those forecasts should be, and you can help me get it to the people at the ground level who are actually going to benefit from more information to make decisions about their own careers and their lives. That's one thing I want to get out of it.

But there's another thing I want to say about this, which is that one of the things that's opened my eyes at OpenAI is that it's not just about technology replacing jobs or augmenting jobs. There are many parts of work that don't have

have enough humans to do them. So if you think about providing mental health services or agricultural extension support systems in emerging markets or small business consulting, all around the world, these are things that we know have really high ROIs.

There are people who can't participate in the job market because they're facing a mental health challenge. There are farmers who cannot get enough yield out of their land because they don't know what seeds and fertilizer to use. There are small businesses that cannot grow because they don't have the kind of advice only a big management consulting firm can afford.

It's those areas I don't want us to lose sight of, and it's such an important part of many of the organizations here. We can scale intelligence to bring the best of what humans know to more people. So I just ask us as we look through this conversation to focus on.

both helping people who are going to be affected by the transition that AI is going to bring, but also unleashing AI to extend the power of intelligence to more people. I think if we do those things, and I think the organizations here are working on both kinds of problems, we're going to have a brighter future together. So that's how I've been thinking about it. I hope you'll join me in that as we think about this.

I now want to call Rose and Jared up to the stage, and maybe I'll pick up the mic here and we'll introduce them. Come on up. Good to see you guys. And I'm going to ask, there we go.

And I'm going to ask these two amazing leaders who are really coming back to share what they've learned as I understand. And I'm going to ask these two amazing leaders who are really coming back to share what they've

right, because they're the OGs, they're from the old cohort, right, they're the ones who have been successful, gone through this process, but have a lot to share, and I think both of them are running organizations that are really sort of embracing amazing work and executing on amazing work at a time of a lot of transition, and so I want to hear and learn from that experience.

Rose, to start, just tell us about yourself, and also the problem that your organization is trying to solve.

Well, first I just need to say thank you so much, GitLab, and also OpenAI. I know you just asked me about challenges, so I just want to really reinforce what you just shared about the opportunities that AI has to unlock.

problems for smaller organizations. I will say a lot of the challenges that we're facing at mRelief have been made even more solvable through once-in-a-generation technology like chat GPT and a lot of the incredible technology that's coming out of this building.

So my name is Rose Afriyie and I'm one of the co-founders of the non-profit organization mRelief and basically what we do is our mission is to transform access to social services for the inherent dignity of all people.

We're addressing the $13 billion in unclaimed SNAP benefits that go unclaimed each year due to a few different challenges. Some of it has to do with

the lack of information, and also we see cumbersomeness in the ways in which people are engaging with government websites, legacy systems, et cetera. And so we've built this easy-to-use tool online and also on text messaging because of the digital divide to help people simply sign up for Snap with ease and dignity.

And I'll share in a little bit how we're leveraging AI in this work, but we have one of the first personalized chatbots to help people end-to-end with the enrollment process, and thanks to GitLab, we've done a lot with respect to the required verifications that need to be submitted for the Snap enrollment process.

Rosemary.
Relief is amazing. This one just like popped off the page when I read about it because there's a lot of work in economics about the value of information to people making decisions. That's what's informed a lot of my work here around forecasts, but what people really need, whether it's the FAFSA or registering for benefits that they have earned and deserve, is navigation and support, and I think the more we can do these kinds of efforts and also evaluate them, which I think you're interested in as well in doing, I think we can scale them to so many people.
So this is exciting. We want to hear more about this.

Jared, how about you? Tell us where you fit in. The workforce development piece is sort of the other end of the coin, right? If people need to get these benefits, they also need to get trained for the jobs of the future.

How is CareerVillage working on that?

Absolutely.
So CareerVillage is...

It is a nonprofit organization. We've been around for 14 years, and what we do is help people navigate the changing labor market. We do that directly to everyday knowledge workers who are trying to get help or workers in any occupation.

We also do that by providing services directly to the skills training organizations who are helping folks get ready for the labor market, and we do that in two ways. You do it using the power of people with a very large online community where young people can ask any question about any career and get advice from, at this point, over 200,000 working professionals who've signed up to give advice.

It's an amazing service, organization, a great community, and that's available at careervillage.org.

Ever since five to six weeks after OpenAI dropped ChatGPT back in November of 2022, we started working on Coach, the AI Career Coach, which you can access at aicareercoach.org.

That has been a tremendous game changer. Now we have the power of people. We've got the power of AI, and increasingly, people want to be accessing both at the right times for the right purposes.

It's really been a tremendous game changer for helping people navigate the tremendous change that's underway or that is coming down the pipe, really, at the end of the day.

end of the day, agility is critical at this time. When you have change, you've got to help labor be agile in navigating that change.

Yeah, I mean, Jared, one thing about your organization that stuck out to me was 14 years you've been at it, right? In Silicon Valley, sometimes we think newer is better, right? Start a company this year to solve this big generational problem. But those of you who've been at it for a long time, you actually have some of the institutional history and knowledge of the transitions that have come before, which could actually really inform the different things you're doing today. And you're building on that capacity, which I think is interesting and was incredibly useful for me.

I also think, look, the hardest question I get for my kids or anyone who's a parent is, what should young people

study in school? How do they prepare for the jobs of the future? And having data driven insights around that is incredibly valuable. I want to double click on a couple of these things going forward.

You know, Rose, we talked a lot about data. And you and I have shared some e-mails and shared some data back and forth. Share with the audience one data point that you really feel you're the most proud of that exemplifies the mission of your organization.

Well, I think when I think about data points, I think about the incredible support that we've been already able to get from GitLab and the investment we made in verifications and submitting all of the correct verifications the first time around because of the bet that GitLab has provided.

Thank you.

lab made on Emory Leaf, we were able to increase that by 20% in the state of South Carolina. A huge, huge improvement, in part because what it means that you can get all of your verifications in is that you will not be delayed one more day in not being able to meet your critical food needs. And so I'm just so proud of that, and it's exciting that we're now laying the groundwork so that we can scale that beyond the 5.2 million people that have already unlocked $2.1 billion in SNAP benefits. Wow.

I think about this. You know, you all are following this debate and discussion around agentic AI, right? Agentic sounds like something I will use to book a plane ticket more easily.

It can do that but agents can also navigate bureaucracies for you
Which I find to be really interesting and they get opportunity
particularly in emerging markets in the United States
Tremendous opportunity to cut through red tape with agents
and I think you're an example that already
and I want to talk more about the potential
To scale that

Jared for you a lot of data around sort of career village as well
Tell me what data point you're most proud of and what it means.

Yeah, I think I'll probably Cheat and use a couple but just from a scale standpoint
I think our whole team is incredibly proud that we've been able to engage over 10 million people
and giving or getting career advice from our various services
Which I think is speaks to the ubiquity of the need for folks to get support with navigating the labor market

and we have very high expectations for how that number is going to continue to ramp up over the coming couple of years. I think I'd speak on the AI component. I think it's worth noting for the non-profits in the room or those who are increasing their investment in AI, we go through some eras. At the very beginning we were trying to figure out if there was demand for such a service. Got it. Validate it. Then we've got to figure out if we can safely and responsibly deploy it. Nailed it. Then you move into can we get an engaging and a high quality experience? Nailed that too.

Tremendous investment in quality assessment is way more important than just can you make something that actually can interact with people.

Okay, but now we're at the phase where we're actually measuring what the results are and I have to tell you that is the most fun part of all. So we have this parade of data coming in over the coming several months throughout this academic year which we're really excited about.

The very first earliest versions of that are coming in now where, for example, in a job training program focused on getting people certified. In a controlled setting, those who are using the AI career coach are more likely to get certified and get more certifications than in a matched group that doesn't have access to it. I think that is just early leading indicators.

of a lot of data that's gonna be coming down the pipe over the coming months. So I'm really, really excited for this phase, after you go through all those early eras, the phase where we get to say, is it actually helping people get into those jobs quicker?

That's fantastic. I mean, this is our vision too, about how you're gonna develop evidence to figure out what works and then scale it. And I know a lot of the funders in the room are thinking about that in the same way.

Rose, I mean, you know, there are a lot of ways that this story is gonna end, right? One is where we do everything it takes to make sure that AI is a leveler and expands opportunity for folks. And another one where we fall short. Talk a little bit about sort of what the most positive vision you see coming out of AI, where you see a transforming opportunity.

And maybe talk to us about what keeps you up at night, because honestly, there's a lot of fear and anxiety about new technologies, particularly one as powerful as AI. And it'd be good to hear from folks, from someone like you who's leading the field, kind of what's worrying you right now about how it's being rolled out.

So I love this question, and I love how we're balancing transformative possibility alongside limitations and risk, because it's always important to balance that. We hear a lot in the field about limitations and risk, and we have to really embrace a holistic mindset.

So first, I wanna talk about what makes an organization like M Relief true. How does that even come into existence? So I.

I co-founded M Relief with my incredible co-founder, Genevieve Nielsen, who's in the crowd today. And we did it after we met in coding school, right? And that was an expensive thing that we did together. It takes several thousands of dollars to be able to go to coding school, at least circa 10 years ago, 11 years ago now, totally, when M Relief was first founded.

And I think one of the great things about AI today is that AI can be a tremendous resource to level up people who have the desire to solve a problem in their society, right? And I think today it's much easier to learn how to code to start something because of.

things like chat GPT. And I'm really excited about the possibilities for there to be as many as 100 M Reliefs or 100 organizations more that are thinking about critical problems of the day.

I also think, when I think about our field and economic opportunity, I think about another very specific individual that's typically in the loop when we think about software, which is social workers.

I went to school. I did my MPP at University of Michigan. And with a lot of folks who pursue their MSW, they were excited about trying to help people get to their heart and soul work that was also at the intersection of economic independence. But when they actually got into the job.

market, what are they working on? A lot of rote, repetitive tasks that are so administrative in nature that it really disallows them to really focus on their core vocation. And I think the real possibility of AI is how can we unleash this workforce, really be an enabler for some of their best talents and gifts, and really help them with some of the coaching work even that I think really needs to be true so that we can get people to that zen in terms of their potential, that highest potential.

But of course, there are things that absolutely keep me up at night. So I would say one of the key things I think about is the digital divide and the linguistic divide.

a stick divide, right? So for us, one data point that I come to a lot is that 21% of people who make under $30,000 a year, according to the Pew Research Center, do not have access to a smartphone, right? So how can we engineer solutions, and I know EmRelief does that a lot, that are really focused on this core demographic that need to hear from us over text messaging or over a phone conversation, right?

So voice is something we think about a lot. And then, of course, I think about the absence of speech corpora or text corpora for the true languages that represent the diversity of the United States of America, right? That's something that I'm really thinking about. Of course.

course, as a nonprofit founder, I also think about trends across philanthropy. There's tremendous investment right now in terms of upskilling for AI, which I think is correct and it's important. But I also know that you cannot be upskilled on anything unless you can also meet your basic food needs. So I think really being able to think about holistic investments across, yes, upskilling, but also the safety net so that we can have this more balanced approach of making sure those interventions that come through for upskilling are actually also effective. So that's what I'm thinking about.

Now, Rose, I mean, you've given us a lot to think about. I mean, this is kind of part of the vision we have here, which is there are so many people who are burdened by all this paperwork, which is really

structured data, which, you know, General Artificial Intelligence is pretty darn good at, handling structured data. And when you think about whether it's a social worker dealing with filing her reports, or a, you know, a forensic scientist who's trying to clear things that maybe the prosecutor's already dropped the case, that happens in a lot of crime labs, for example, where they're working on things that aren't relevant anymore or you're thinking about a small business owner trying to navigate a tax credit that they deserve. These are all things that we could use AI to help sort of clear the desk and allow them to focus on the more valuable parts of their job to them and also drive higher productivity in all those use cases. But we're not gonna do it unless we make sure that this technology is in the hands of more people and that it's linguistically relevant, to your point. You know, one of the cool things.

things at OpenAI that I've admired is our commitment on all the different languages, so a lot of governments are using us for translations, state governments here, so they can translate services into lots of different languages, and then, of course, 1-800-CHAT-GPT, right, which when they first started, I wasn't aware of the stat you talked about. I'd probably take it for granted, but oh, my gosh, it's interesting to see people taking that up and accessing CHAT-GPT through a different mechanism. So we can do more here, and that's definitely an area we've got to work with you and others on.

Gerard, I'll ask you the same question. What's that positive vision out there? What should we be shooting for? And taking Rose's really excellent point, I mean, what are the risks we should be attending to as well?

Yeah, I think of it differently on the supply side or labor versus on the demand side, which

employers. On the employer side, I think there is a very natural opportunity that's going to be very present for a lot of firms to automate existing work. It can happen. It should happen. It's a normal part of the course of a transformative technology.

What I think is going to be really important on the demand side, however, is to make sure that they're actually thinking longer term. I'll just tell you right now, it's better for longer term shareholder value to create new services and lines of business with a transformative technology.

What I'd like to see on the demand side is that they're doing both, that they're not just working on automating business processes, but they are also at the same time leveraging AI to make sure that they're not just working on automating business processes, but they're

do something really new in the marketplace. That's going to end up creating new demand for new skills, new work, new talent that's going to create that sort of paired, and I think probably to your point earlier where you were asking the question, Ronnie, of is this change going to be like previous technological changes?

I think one of the most important mediating factors is going to be the extent to which firms are using this technology for new value creation and not just cost reduction.

On the supply side, however, I think the most important thing, and I mentioned it before, is agility. It's both the opportunity, because individuals are extremely agile, and when presented with an opportunity that they understand and can see

They will go and learn that skill. They will want to get that job. But the institutions is what keeps me up at night. How much agility is there in the education system? In the higher education system? In the workforce development system? And a lot of the most agile institutions are coming to institutions like ours trying to get help because they want to be agile. We've really got to support them and we've got to say this is not business as usual. This is a transformative moment in the labor market and we need to dramatically increase agility in the education, higher education and job training ecosystems.

with some business leaders. We do a lot of work like this to try to emphasize that long-term vision around value creation. That's something that we can do.

On the second piece, in terms of working together to make sure the institutions are up for the task, that's where we actually need to work together a lot, I think, right?

Because we all have different touch points for this, but it's not something we can do alone or you can do alone, and this is something I think we should take away from the conversation.

Let me ask you guys another question about this in terms of, I mean, the learnings that you think folks in the audience should take.

In addition to doing this amazing work and talking about the people you serve, you're also organizational leaders. You've gone through this process before, you've been successful.

If I'm sitting in the audience today running my own organization, Rose and then Jared, what advice do you have to the folks in the audience about how they should approach some of these challenges?

Well, I think there's a few things. I think one, now is the time to be building with AI. I just really think in our field, I see acute time poverty standing in the way of really being able to achieve what needs to be achieved so that we can live in service of our beneficiaries and also achieve our missions.

And so that's part of the backdrop. The other part of the backdrop is that we're seeing drastic cuts across the safety net, drastic cuts across education, and it's really important for us to take that seriously and think about ways that we can innovate to really step in the gap and provide information for people when they most need it.

So that would be first. The second thing is I have a genuine belief that we are experts of our own experience and there's a lot of expertise that you've already been able to amass in your own respective fields.

Lift while you climb. I think never forgetting that once you've learned something new, once you've been able to master something, being able to reach across the aisle, across the organizations that you support or that you're in communities of practice with to really be able to do that.

I know myself, I've been able to co-create frameworks that have been backed by Microsoft just on AI and non-profit leadership. And that's something that I've been committed to doing and I'm still committed to doing. So that's just something that I also want to encourage more folks.

folks to do. And then, of course, I just have to say, and because agility has been mentioned, I want to amplify Jared's mention of it and Ellie's mention of it, I think in philanthropy we need to also be able to support the kind of distribution of capital that allows for folks to learn and be transparent about their learnings, and I'm just so glad that that's been a part of our support from GitLab. And I just want to challenge more folks who are in philanthropy to, yes, be focused on outcomes, it's important, but also really support capital that makes it safe for people to share learning.

Just one follow-up, because I think if I was in the audience, I'd be saying, okay, I heard her message loud and clear, now is the time to build with AI, but then I'd say, man, I

I'm not Rose, I can't do this, how should the rest of us make time with all the other things we're doing running this organization at a time when resources are scarce, when people are nervous about the future, how do we make time for this stuff?

I think you're saying something that everyone wants to internalize, but what can they do? Anything that they can do right now at their jobs to incorporate this, because I think a lot of people just feel overwhelmed, how do you think about that piece?

Yeah, I mean, I think about even just like the small things that you can leverage AI to do, whether it is your strategic planning process. I know as a non-profit leader, we just came out of strategic planning just last week, how can we leverage AI to get as quickly as possible to our work plans, our action planning, right?

because we know we can sit in those rooms and have that conversation for days and days and days and days and days, right? And that process by which we really metabolize those learnings and make them actionable can be significantly accelerated through things like AI. So even if it's just like you're starting and it's internal uses, I think, that you can get the most traction with first before we're even thinking about external uses. That's what I'm saying.

I love it, I love it. One of the nonprofit leaders I admire, he said, pick the most annoying task that your organization, pick that one, right? Pick the most annoying, hardest one and put some chat GPT on it, right? And see what it can do. And honestly, what you got to lose, right? It's already the most annoying thing you're doing, right? So you can't really make it much worse. And a lot of times.

you might find out that there's a hack. Strategic planning is a great one.

Jared, same question for you. What advice do you have for folks in the audience who are trying to balance these things? They want to adopt AI. They're using it in very innovative ways if they're in this room. Any lessons kind of looking back now?

Yeah, I think our team has gone through a real transformation as an organization. Just getting deeper and closer to the technology. It has been a long journey. A lot of people have had to upskill themselves.

I think one of the things that I've really been inspired by is the way that our non-technical colleagues have been really leaning into and adopting this technology. I think it's really coming to a head with

the transition to agentic AI, where we are not just building real agency into the AI career coach, so the bot itself, which is gonna be much more proactive, is gonna be finding new opportunities for people in a way that's really valuable for them,

but the staff, so for example, for our agentic AI transformation of our operations, we chose a enthusiastic, non-technical team to be the first ones to start taking tasks that they, as you said, hate doing, right, but need to do, they don't feel value added to them, automating those and trying to figure out what's the right configuration, what's the right opportunity, what's the right set of systems.

to understand our customers better and what they want because we can see the big picture and pull everyone in to make it successful.

It's like a feedback loop.

That's a really good point. It's really exciting when we can bring that data back into our models and make them even better as a result.

Exactly, yeah. Great collaboration.

Yeah, great work, team.

What is gonna be the challenge that employers are gonna face in trying to do that? We need to know how, the rate at which, you said this actually a couple weeks ago in an OpenAI forum session where you had talked about how the rate at which firms are gonna be adopting, and when you're talking to CIOs and other leaders, the challenges that they're facing and having to roll out this technology at large scale, we need to understand the extent of the friction that they're gonna face, and I think we're doing that firsthand with our non-technical team, so I would recommend that.

That's really smart. I think sort of non-profit leaders are probably really well positioned for this. Private sector leaders sometimes overlook the non-technical people on their team, and when they roll out.

something like Chachi VT, I've heard this story time and time again, how it's the least likely people in the organization, and you can think about how one might define least likely, but least likely, less technical, who are the ones who actually get the most out of it and sometimes teach everyone how to use it.

In your organizations, right, you're finding the same, and I think that's actually a lesson that can go from the nonprofit to the private sector in a big way. That's what we see in implementations all the time.

I'll also add leadership matters, like when I get on these calls with enterprise leaders, if I don't see that the CTO and the CIO are really bought in to these enterprise implementations and really have change management in mind, rather than just a point solution, dropping in AI and expecting magic, it's probably unlikely to succeed.

I think you're going to see these things in the non-profit case as well, so leaders like this and in the audience really matter.

One last question for you all, and then we'll wrap up. How do we make sure what we're learning in our own non-profits gets disseminated to others? There are so many non-profits. In my hometown, there's 11,000 non-profits, and I wonder, how are they all going to do interesting stuff and learn from each other so we don't have to reinvent the wheel?

Can I get quick answers from Rose and Jared, and then we can wrap up on that?

Yeah. I think, first, it's events like these, just so that we can really be able to have that. But I also think it's about having the support and the capital so that you can learn in public.

and share what you're picking up on the ground. I think that's just so, so critical. I know for us at Emory Leaf, we're thinking about things like voice and the transformative potential for that to help with application processes with, of course, community-based organizations in the loop, but that's so critical.

And then also, the potential for AI to be a buffer between people who are going through really challenging, challenging life problems, whether it be setbacks in career, or even challenges with respect to a diagnosis that really alters your economic outcomes. I think the idea that AI can be something that steps in to be able to have the conversations.

with state governments or have the conversations with case workers who might be there on obligation and not necessarily true vocation. I think the idea of having some of those learnings at the table is something that we're really committed to do so that our families can just thrive.

Yeah, and I think I'd say, so I think definitely, Rose, I totally agree with you. Opportunities like this are essential. Accelerator programs are valuable. Communities like the Open AI Forum, like Fast Forward, these things are essential to giving us the opportunity for us to be able to come together to compare notes. When that opportunity arises, the most important thing is to show, not tell.

How are you running evals, I want to see it. Oh, you have human in the loop? That's great, show it to me.

I think that, and I'll show you mine. I'll show you mine. And I think that allows us to quickly eclipse what the private sector is doing because we're so open with each other.

And I think if you just show, don't tell, we uplift each other.

Well, with that, I just want to thank you two for uplifting us today. Rose Afriyie and Jared Chung, I couldn't have imagined two better leaders to have on the stage.

So thank you so much for being on the stand.

Thank you. Thank you so much, appreciate it.